I think this behavior is generic, and I can see reasons for preserving
it. You might want to access the estimates of a program that failed
to converge, e.g.
webuse nlswork
tsset id year
xtgee ln_w union ten grade south not_smsa, corr(uns)
mat li e(b)
mat li e(V)
but it would be nice if there were a common return code. The -xtgee-
example returns 430 in e(rc) --see
search r(430)
but -qreg- returns e(convcode) etc. It would be easier to trap errors
if there were a generic e(anyerrors) or somesuch. As it is, it falls
to you to anticipate and trap all the appropriate errors.
On 10/26/07, Ricardo Ovaldia <[email protected]> wrote:
> Thank you Jeff. I assumed that was the case. However,
> should the -e()- results be clear if the regression
> fails?
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
